According to the Webster dictionary anything that uses a scientific method could be said to be a science. A scientific method aims to observe and experiment physical data to reach conclusions which either confirm or nullify a theory. However it can be argued that not all of psychology can follow strict scientific methods and in turn be called a science. For example the research of personality can be complicated by ethical issues; a researcher may wish to perform brain surgery to test the effect that removing parts of the brain has on mood and personality but of course this is both illegal and unethical. Therefore psychologists tend to perform on animals and extrapolate the results to humans. One way to get around this maybe retrospective studies which attempt to draw conclusions from past events rather than setting up a strict scientific experiment in the present. For example finding a corpse and establishing a cause of death. The biggest problem is acquiring the information for example with the risk of cigarette smoking well publicised lung cancer patients are more likely to remember their previous smoking history then the controls would. It can be argued that retrospective studies can not reliabily distinguish between cause and effect and any conclusions drawn from them are suspect.
If you strike 10 pool balls with a stick in the same way under the same conditions you can calculate in advance what the reactions will be. However poke ten people with a stick and each of their reactions will most certainly be different. Nonetheless psychology can produce objective evidence that can be replicated with the same success as experiments in physics and chemistry.
A theory allows one to make prediction about observation not yet made. The predication allows for a laboratory test that might support or falsify the underlying theory. The idea of a paradigm and falsifiability are very important. A paradigm in this sense is a shared set of beliefs. Falsifiability on...